Targeted advertising to certain demographic groups based on data-mined aggregated data can deliver positive ad impressions of products, services, and brands. These advertisements (“ads”) generally target specific demographic groups within a population to which the subject matter of the ads are predicted to be statistically relevant. Often times these targeted ads are delivered to specific targeted users of social networking sites, such as Facebook or Twitter.
However, data mining techniques are limited in that they cannot access information that individuals have not shared or otherwise supplied as part of the data being mined. Such information might include sensitive or confidential information, for instance. Such information, referred to herein as a user's “offline backstory” or “backstory”, is present for virtually all social network users, though they tend not to explicitly post such information on social network sites for obvious reasons. Accordingly, certain ads that are statistically likely to be useful or positively welcomed by many users of a particular demographic may in fact leave a negative impression with a targeted user of that demographic when the statistical predictions are at odds with the user's backstory.